EP2988124A2 - Scalable and high throughput biosensing platform - Google Patents
Scalable and high throughput biosensing platform Download PDFInfo
- Publication number
- EP2988124A2 EP2988124A2 EP15184108.7A EP15184108A EP2988124A2 EP 2988124 A2 EP2988124 A2 EP 2988124A2 EP 15184108 A EP15184108 A EP 15184108A EP 2988124 A2 EP2988124 A2 EP 2988124A2
- Authority
- EP
- European Patent Office
- Prior art keywords
- biochip
- sensors
- signal
- sensor data
- interface
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/483—Physical analysis of biological material
- G01N33/487—Physical analysis of biological material of liquid biological material
- G01N33/48785—Electrical and electronic details of measuring devices for physical analysis of liquid biological material not specific to a particular test method, e.g. user interface or power supply
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/72—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables
- G01N27/74—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables of fluids
- G01N27/745—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating magnetic variables of fluids for detecting magnetic beads used in biochemical assays
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/0023—Electronic aspects, e.g. circuits for stimulation, evaluation, control; Treating the measured signals; calibration
- G01R33/0035—Calibration of single magnetic sensors, e.g. integrated calibration
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/12—Measuring magnetic properties of articles or specimens of solids or fluids
- G01R33/1269—Measuring magnetic properties of articles or specimens of solids or fluids of molecules labeled with magnetic beads
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R33/00—Arrangements or instruments for measuring magnetic variables
- G01R33/0094—Sensor arrays
Definitions
- the present disclosure relates in general to biological analysis, and in particular to scalable, multi-channel high performance biosensing platforms, which are capable of high throughput biological analysis.
- biochip-based point-of-care devices Conventional biological analysis methods are being gradually replaced by sophisticated and compact biochip-based point-of-care devices. These biochip-based point-of-care devices perform common blood or glucose analysis at a fraction of the cost of conventional biological analysis methods. Also, biochip-based point-of-care devices typically require a much shorter analysis time compared to conventional biological analysis methods. Given the high potential of these biochip-based point-of-care technologies, research in biodetection methods and in biochips themselves, has significantly grown in the past decade.
- a biological detection system is provided.
- the biological detection system is characterized by a biochip and a biosensing platform.
- the biochip has at least two blocks of sensors, each block of sensors having a signal acquisition channel that is independently accessible such that each block of sensors can be read in parallel.
- the biosensing platform has drive circuitry to drive each block of sensors of the biochip.
- the same or different driving signals can be applied to each block of sensors of the biochip.
- the biosensing platform further has a set of acquisition interfaces arranged such that a select one acquisition interface is coupled to a corresponding one signal acquisition channel of the biochip.
- the biosensing platform also has a set of co-processors such that a select co-processor is uniquely associated with a corresponding one of the acquisition interfaces. Still further, the biosensing platform has a central controller that controls the drive circuitry. The central controller also interfaces with the set of acquisition interfaces. Still further, the central controller interfaces with the set of co-processors to read data from the biochip. Moreover, the biosensing platform may provide a direct user interface, a remote interface or combination thereof.
- a method of operating a biological detection system having a biochip may be used, for instance, to calibrate drive circuitry to achieve a desired signal to noise ratio of signal data read from the biochip.
- the method is characterized by setting drive circuitry to generate a set of drive signals and by reading sensor data from the biochip.
- the method is also characterized by performing digital processing with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip.
- the method is further characterized by using the result of the digital processing with a low frequency resolution to define a preliminary indication of a target frequency to measure a sensor data output from the signal acquisition channel, and by identifying a range or band of target frequencies.
- the method is characterized by performing filtering and signal demodulation to displace a frequency (band) of interest from the identified range of target frequencies down to a base band and by designating a bit and frequency resolution from the base band that maximizes the signal to noise ratio of the data read from the signal acquisition channel.
- the method may be further characterized by using a modulation technique to drive the biochip and correspondingly, performing demodulation to extract sensor data from the biochip.
- performing digital processing is further characterized by performing a fast Fourier transform with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip.
- digital signal processing may be implemented for performing one or more complex correction algorithms, which may include a Kalman filter, drift compensation, and a temperature controller, on the result from the Fourier transform.
- systems, methods, and computer implemented processes are utilized to implement scalable, multi-channel high performance biosensing platforms, which are capable of high throughput biological analysis.
- aspects of the present disclosure find use in applications such as gene expression profiling, where thousands of biosensors analyze a given sample in parallel.
- the biosensing platform described herein integrates the complexity and computational needs of the electronic sensing system, into portable and standalone embedded systems with adequate computational capabilities for such high throughput analysis.
- systems are provided to deliver the computational power required by advanced biochip technology, and to allow simultaneous acquisition and processing at high sampling rates, through a high-performance and configurable acquisition and processing architecture.
- a biosensing platform is provided, which uses multiple dedicated co-processors performing Fast Fourier Transform (FFT)-based signal filtering, as well as other complex correction algorithms.
- FFT Fast Fourier Transform
- An additional dedicated processor is used to synthesize the stimulus signals for the sensors, and a central processor is provided to control the whole system.
- the biological detection system 100 includes a biochip 102 and a biosensing platform 104.
- the biological detection system 100 also integrates with an interface 106 to provide a complete lab-on-chip solution.
- the interface 106 may include for instance, local user interface peripherals (such as a keyboard, mouse, display, etc.), a remote interface, or a combination thereof, which can exchange data with electronics of the biosensing platform 104.
- the biochip 102 (e.g., a magnetoresistive biochip) includes one or more "blocks" 108 of sensors. For instance, as illustrated, there are n blocks of sensors where n is any integer greater than zero. In this manner, particular blocks are referenced as 108-1, 108-2, ..., 108- n , etc. Each block 108 of sensors has a signal acquisition channel 110 that may be independently accessible such that each block 108 of sensors can be read in parallel, as will be described in greater detail herein.
- the biochip sensors (e.g., the blocks 108 of sensors) achieve a high sensitivity, high Signal-to-Noise Ratio (SNR) and high dynamic range when used in conjunction with the biosensing platform 104 described herein.
- SNR Signal-to-Noise Ratio
- each block 108 includes a plurality of sensors.
- Each of the plurality of sensors can be of a various type (e.g. magnetic, temperature).
- there are 256 sensors in each block 108 arranged in an array of 16 rows and 16 columns.
- more than 256 sensors are required for many important applications (e.g., gene expression profiling).
- the illustrative biochip 102 combines CMOS with thin film fabrication technologies to provide the required number of sensors on a small biochip package.
- each block 108 can have any number of sensors in its array.
- each block 108 can include 16 sensors or more in certain illustrative implementations.
- each block 108 of the biochip 102 may have a single acquisition channel.
- Each acquisition channel (or block of sensors themselves) can include integrated on-chip signal conditioning circuitry, and necessary readout circuitry to address and access the individual sensors of a corresponding block 108.
- a sensor reading technique such as Time-Division Multiplexing (TDM)
- TDM Time-Division Multiplexing
- each sensor can be read out individually using time-multiplexed reading techniques, such as by using a first multiplexer ("row multiplexer”) to select a desired row, and a second multiplexer (“column multiplexer”) to select a desired column.
- the sensor that is read out through the signal acquisition channel is thus the sensor at the intersection of the row selected by the row multiplexer and the column selected by the column multiplexer.
- the biochip 102 is characterized by each block of sensors having an array of at least 16 sensors and a corresponding time division multiplexer to read out individual sensors from a corresponding block.
- the biochip 102 can provide a multichannel capability that includes up to, and/or in excess of 1000 sensors in certain exemplary implementations.
- the biosensing platform 104 includes in general, drive circuitry 112 that generates the necessary drive signal(s) for the corresponding biochip 102. More particularly, the biosensing platform 104 includes drive circuitry 112 to drive each block 108 of sensors of the biochip 102. In this regard, the manner in which the drive signals from the biosensing platform 104 couple to the biochip 102 will depend upon the particular implementation of the biochip 102. However, by way of example, the drive circuitry may include circuitry to generate bias currents and oscillating magnetic fields that are applied to the sensors of the biochip 102.
- the drive circuitry 112 can include a single circuit with multiple outputs.
- the drive circuitry can include multiple independent circuits, e.g., an independent circuit for each block 108 of the corresponding biochip 102.
- the one or more circuits that make up the drive circuitry 112 can be software-based, hardware based, or a combination of the above.
- drive signals can be fully or partially programmable, controlled by set points, parameters or other program variables, made adjustable by control signals, e.g., programmed by control voltages or other control signals, etc.
- the drive circuitry 112 is controlled by a central processor 114.
- the drive circuitry 112 is controlled by the central processor 114 to have an adjustable frequency.
- the central processor 122 can control the drive circuitry to generate arbitrary signal shapes and amplitudes.
- the drive circuitry 112 can utilize Direct Digital Synthesizers (DDSs) that can be combined with arbitrary signal generators deployed in a reconfigurable fabric, which are controlled by the central processor 114, to provide the correct signals to the corresponding biochip 102.
- DDSs Direct Digital Synthesizers
- the biosensing platform 104 also includes multiple (e.g., a set of) acquisition interfaces 116. As illustrated, the biosensing platform 104 includes n acquisition interfaces 116 where n is any integer greater than zero. In the illustrated example, a select one acquisition interface 116 is coupled to a corresponding one signal acquisition channel 110 of each block 108 of the biochip 102. Thus, where the biochip 102 has 4 blocks 108, each block having a single, signal acquisition channel 110, the corresponding biosensing platform 104 includes 4 acquisition interfaces 116, each acquisition interface 116 connected to a corresponding unique one of the biochip signal acquisition channels 110.
- each acquisition interface 116 includes a converter 118, e.g., an analog to digital converter, which receives the signal from its corresponding signal acquisition channel 110.
- the analog to digital converter converts the sensor signal into a digital representation that can be further processed, as will be described in greater detail herein.
- an acquisition interface 116 is configured to measure sensor data from the corresponding signal acquisition channel 110 at a specific binding frequency or frequency band that is determined by the modulation induced by the magnetic field, the bias current, the sensor's response to these AC signals, and the system's noise and cross-talk.
- each analog to digital converter 118 may have a sampling frequency of at least 1 kHz and at least 12-bit resolution to accomplish the desired sampling.
- the output of the analog to digital converter 118 couples to a controller 120, e.g., a serial peripheral interface controller.
- a controller 120 e.g., a serial peripheral interface controller.
- Each acquisition interface 116 may also include signal conditioning circuitry as necessary to interface with the analog signals from the biosensor 102.
- the controllers 120 are each controlled by the central processor 114.
- each controller 120 couples to a corresponding co-processor 122.
- each controller 120 conveys information from its corresponding analog to digital converter 118 to its corresponding co-processor 122.
- the illustrated biosensing platform 104 includes a set of co-processors 122 (e.g., n co-processors where n is any integer greater than zero) such that a select co-processor 122 is uniquely associated with a corresponding one of the acquisition interfaces 116.
- a single co-processor 122 can be utilized to service a subset of acquisition interfaces 116.
- Each dedicated co-processor 122 performs digital processing, e.g., Fast Fourier Transform (FFT)-based signal filtering on the data from the corresponding acquisition interface.
- Each co-processor 122 can also/alternatively perform other complex correction algorithms, e.g., a complex correction algorithm including a Kalman filter.
- each co-processor 122 can be used to implement a temperature controller, a complex modulator and demodulator, etc.
- the co-processors 122 also communicate with the central processor 114, which controls the biosensing platform 104.
- the central processor 114 of the biosensing platform 104 controls the drive circuitry 112.
- the central processor 114 also interfaces with the set of acquisition interfaces 116 (e.g., via the controllers 120). Yet further, the central processor 114 interfaces with the set of co-processors 122. Also, the central processor 114, a dedicated co-processor 122 or other processor (not shown) can be utilized to synthesize the stimulus signals for the data read from the sensors of the biochip 102.
- the illustrated biosensing platform 104 further includes a main memory 124.
- the main memory can couple to the central processor 114 and/or to one or more of the co-processors 122.
- the central processor 114 and the co-processors 122 can share the main memory 124.
- main memory 124 may be used to store programming, control information, calibration information, data, etc.
- a computer processing device e.g., personal computer or server computer
- the biosensing platform 104 can be coupled to the computer processing device, e.g., to download collected signal data, etc.
- the biosensing platform 104 includes electronic circuitry for addressing the biochip 102, for driving control signals and for conditioning and acquiring the biochip signals.
- the acquired signals are processed by a set of co-processors 122 and are transmitted to the central processor 114, e.g., which executes software that functions as a digital analyzer where, through the interface 106, e.g., a conventional graphical user-interface, the experiment is controlled by the operator or analyst.
- the biosensing platform 104 provides multiple signal acquisition channels 110 operating at a high sampling rate, thus producing a high volume of data.
- the biosensing platform 104 also includes the necessary computation power that is required to handle the volume of data.
- the interface 106 may comprise a monitor, keyboard, mouse or other computer peripherals, a graphical user interface, network interface, etc.
- the biological detection system includes a biosensing platform implemented in two components, including a programmable logic component and a processing system component, wherein the drive circuitry 112 (e.g., drive signal synthesizer), the co-processors 122 and the acquisition interfaces 116 are implemented in the programmable logic component, and the central processor 114 and the main memory 124 are implemented in the processing system.
- the drive circuitry 112 e.g., drive signal synthesizer
- the co-processors 122 and the acquisition interfaces 116 are implemented in the programmable logic component
- the central processor 114 and the main memory 124 are implemented in the processing system.
- the central processor 114 may further communicate over a network, which provides communications links between various processing devices, and may be supported by networking components that interconnect the processing devices, including for example, routers, hubs, firewalls, network interfaces, wired or wireless communications links and corresponding interconnections, cellular stations and corresponding cellular conversion technologies, e.g., to convert between cellular and tcp/ip, etc.
- the network(s) may comprise connections using one or more intranets, extranets, local area networks (LAN), wide area networks (WAN), wireless networks (WIFI), the Internet, including the world wide web, cellular and/or other arrangements for enabling communication between the processing devices, in either real time or otherwise, e.g., via time shifting, batch processing, etc.
- a laboratory assay begins by labeling biological targets under analysis with Magnetic Particles (MPs). Then, a bias current, i B , and an oscillating magnetic field signal, h B , are applied to the sensors of the biochip 102, e.g., via the drive circuitry 112 of the biosensing platform 104. Upon biomolecular recognition and washing, the sensors detect the fringe fields created by the MPs that have been captured by biological probes placed above the sensors. The biological information is then extracted by measuring the variations of the sensor voltage signal, v bind .
- MPs Magnetic Particles
- the sensors of each block 108 are measured using Time-Division Multiplexing (other modulation and multiplexing techniques can alternatively be applied). Since reaching a low noise level typically requires a high number of samples per sensor, biochip reading times can be slow, e.g., about 4 minutes. Such biomolecular recognition at a low speed, may be acceptable for certain biochips. However, such slow recognition does not scale when the number of sensors grows to 1000 or more. To circumvent issues of slow speeds, the biochip 102 is organized in groups of sensors, e.g., 256 sensor elements per block 108, and each sensor is time multiplexed within its corresponding block, as noted above.
- each group 108 of the biochip 102 is serviced by a dedicated acquisition channel interface 116 of the biosensing platform 104, e.g., comprising signal conditioning circuits and an Analog-to-Digital Converter (ADC) 118. Furthermore, the data from each channel is processed in parallel, to allow the observation of the assay evolution in real-time.
- ADC Analog-to-Digital Converter
- the applied signals from the drive circuitry 112 include an AC component. Consequently, v bind is measured at a specific binding frequency, f bind , which is determined by the modulation induced by the magnetic field h B ⁇ f 2 ⁇ , the bias current i B ⁇ f 1 ⁇ , the sensor's response to these AC signals, and the system's noise and cross-talk.
- f bind a specific binding frequency
- a flexible and tractable method is to use an FFT of the signal acquired at each of the 1000+ sensors, to compute the signal amplitude v bind at the frequency f bind .
- each co-processor 122 may require a sampling frequency of at least 50 kHz and a frequency resolution as low as 1 Hz, for a 24-bit resolution in each sensor. That said, aspects of the present invention can function suitably, for instance, with a sampling frequency of at least 1 kHz and at least a 12-bit resolution. Regardless, this imposes large memory and hardware requirements, constraining the scalability of the system. Additionally, other complex correction algorithms can be implemented by each co-processor 122, namely but not limited to: drift correction, temperature controller, Kalman filters, complex modulator and/or demodulator (e.g. spread spectrum).
- the computational performance required by the biosensing platform 104 to ensure the resolution and reliability of bio-recognition assays can be achieved by optimizing the processors' architecture and by increasing the exploited parallelism with the usage of multicore structures. Nonetheless, it would be insufficient to exclusively address the performance limitations, since the bottleneck may also arise in the communication channels, such as the one connecting the acquisition circuits to the processing unit. Furthermore, due to the significant complexity of the underlying algorithms, the processing unit must be able to store and manipulate large datasets, making it clear that a high memory bandwidth and capacity are necessary to take full advantage of the enhanced processing capabilities.
- a high-speed serial interface may be selected in order to provide a fast communication between the ADCs 118 and the set of dedicated co-processors 122 (e.g., cores) where the most computational intensive tasks are handled. All components, including the generator of the stimuli signals, are configured by a central processor 114, e.g., also referred to herein as a General Purpose Processor (GPP) which, aside from acting as the main system controller, can also perform auxiliary computations on the acquired data and user interface tasks as well as communication tasks.
- GPP General Purpose Processor
- the central processor 114 shares its main memory 124 with the co-processors 122.
- the central processor 114 e.g., the GPP, is a suitable device to configure the several system components, and thus the central processor 114 plays a key role on easing the system integration.
- OS Operating System
- a processor-memory interconnection subsystem (subsystem that interconnects the central processor 114 to the main memory 124) can be equipped with DMA channels (or equivalent), allowing the central processor 114 to communicate with streaming peripherals within a shared memory space.
- Each acquisition channel is linked to one biochip sensor group 108 and directly feeds one specialized co-processor 122 with the data sampled by the ADC 118.
- each ADC 118 can operate at a maximum rate of 96 Mbit/s, the absence of a high-speed data channel would seriously compromise the overall throughput.
- standard Low-Voltage Differential Signaling (LVDS) was chosen as a cost-effective and noise tolerant solution to achieve high transmission rates.
- SPI Serial Peripheral Interface
- the specialized co-processors 122 are responsible for measuring the amplitude of the sensors binding signal, v bind , from the biosensor 102, at a specific frequency.
- This signal is estimated using a zoom-in FFT approach together with other complex processing algorithms like Kalman filters that improve the overall data accuracy.
- an FFT is performed with a low frequency resolution, and by using a subset of sensors in the same channel. This provides a coarse and preliminary identification of the best f bind frequency to measure the v bind , while keeping the hardware resources and power consumption at tolerable levels.
- band-pass filtering and signal demodulation are performed, to displace the frequency of interest down to base band.
- a reconfigurable field programmable gate array (FPGA) device is regarded as a suitable technology to accommodate the multi-core infrastructure of specialized FFT co-processors, filters and demodulators 122.
- An FPGA combines the required high performance level, while still conferring the flexibility to change or update the adopted processing algorithms.
- multiple configurations can be stored in memory for each zoom-in level.
- the user starts by observing the coarse signal spectrum of the initial subset of sensors. Then, the most favorable demodulation frequency and spectrum band is selected. This will trigger the FPGA reconfiguration and the deployment of the specialized co-processors in hardware, configured with the target demodulation frequency. At this point, the user can observe the binding signal for each sensor.
- a set of I/O peripherals 106 are utilized to make the biosensing platform 104 a standalone analysis and diagnosis system. These modules not only allow the user to take full control of the experiment, but also offer the visualization means for clinical interpretation purposes.
- the standard set of computer peripherals i.e. USB keyboard / touch screen panel and HDMI monitor
- USB keyboard / touch screen panel and HDMI monitor is typically utilized as interfaces to the biosensing platform 104 to enhance the usability experience.
- FIG. 3 a portion of the biological detection system 100 of FIG. 1 is illustrated.
- FIG. 3 illustrates a partial view of the biosensing platform 104.
- a single signal acquisition interface 116 is illustrated for convenience of illustration and discussion.
- the biochip 102 and the interface devices 106 have been omitted for convenience of discussion.
- Reference AC signals must be supplied by the drive generator 112 to the biosensors of the biochip 102, to produce a measurable response (e.g., i B and h B discussed above).
- the amplitude, shape and frequency ranges of these stimuli are partially constrained by the considered biochip technology.
- the noise level that the measurements are exposed to is inversely proportional to the frequency of these signals.
- the accuracy of the results is strongly constrained by this parameter.
- a variable and configurable frequency generator component of the drive circuitry 112 which can be adjusted and parametrized by the central processor 114 (e.g., GPP), ensures the adaptability of the biosensing platform 104 to multiple generations of the biochip 102.
- the required signal generators can be implemented by using Direct Digital Synthesizers (DDSs) deployed in a reconfigurable fabric, which can be combined with additional hardware in the reconfigurable fabric to achieve complex modulation like spread spectrum, and controlled by the central processor 114.
- DDSs Direct Digital Synthesizer
- a Xilinx Zynq 7000 Extensible Processing Platform was utilized to build the biosensing platform 104 as an embedded system.
- This particular System on a Chip includes a state of the art reconfigurable fabric and it offers most of the features required for the system described more fully with regard to FIGS. 1-3 .
- the Zynq SoC integrates two independent modules in the same chip: a Processing System (PS), featuring a processor 132, e.g., a dual-core ARM Cortex-A9 processor, and Programmable Logic (PL), consisting of reconfigurable fabric compatible with Xilinx 7-Series FPGAs.
- a processor 132 e.g., a dual-core ARM Cortex-A9 processor
- PL Programmable Logic
- the PS also integrates a collection of industry-standard communication interfaces 134 (Gigabit Ethernet, USB, I2C and SPI).
- PCAP Processor Configuration Access Port
- the ARM processor is able to exploit the software-driven dynamic partial reconfiguration capabilities of the reconfigurable fabric, thus promoting an adapted and optimized usage of the reconfigurable part of this SoC.
- the PS-PL interconnection can be achieved by using an interface, e.g., the Advanced eXtensible Interface (AXI) bus protocol, through any of the provided connectivity ports, namely, General-Purpose (AXI-GP), High-Performance (AXI-HP), and Accelerator Coherency Port (AXI-ACP).
- AXI-GP General-Purpose
- AXI-HP High-Performance
- AXI-ACP Accelerator Coherency Port
- two master and two slave AXI-GP interfaces 136 connect the PL to a PS central interconnect 138, allowing for low throughput data and control transfers.
- a separate interface 140 e.g., an AXI-HP interface, connects directly to the main memory 142.
- read and write operations are mediated by an asynchronous 1 kB FIFO with a configurable data width of 32 or 64 bits.
- the AXI-ACP interface connects the PL directly to the Snoop Control Unit (SCU), to ensure data coherency at the L2 shared cache of the ARM processors. All these ports can be individually enabled and configured. On the PL side, these interfaces are implemented using the existing reconfigurable logic, thus having an associated cost in terms of hardware resources.
- the reconfigurable fabric on the PL is used to implement the dedicated co-processors and to support the interface with the biochip.
- the base software layer running in the PS is implemented in an exemplary implementation, using the Linaro Ubuntu GNU/Linux OS, which provides optimized tools and software packages for ARM SoCs. By adopting this OS, the PS provides a rich interface to the user by using the existing set of computer peripherals.
- FIG. 3 depicts the block diagram of the implemented architecture, considering a single-channel 116. It corresponds to the initial zoom-in phase, where an FFT is computed over a small subset of sensors from a single acquisition channel to determine the binding frequency, f bind that maximizes the signal to noise ratio.
- the Co-processor channel 1X in Fig. 3 is replicated as necessary, according to the blocks 108/acquisition channels 116 of the biochip 102.
- AXI interfaces connect the reconfigurable fabric to the ARM processor and to the associated main memory, either for control or for high throughput data transfers.
- each devised co-processor provides a slave interface for configuration purposes, along with a master controller that writes in the shared memory bus.
- Data consistency between the dedicated co-processors and the GPP is achieved by using a contiguous region in the main memory, marked as non-cacheable and excluded from the system's memory management.
- the OS is able to read this memory region by remapping the region into the kernel virtual address space, thus allowing it to be easily accessed by a custom device driver.
- the FFT co-processors (e.g., co-processor 122) in the PL are based on the Xilinx FFT IP core v7.1, and adopt a pipeline operation to produce an output per clock cycle in a stream-based processing scheme.
- the cores are configured to receive 24-bit fixed-point real inputs, although a higher precision is used inside the cores to decrease numerical error propagation between intermediate stages.
- embedded DSP blocks are used in the butterfly computations to allow for a better balance in the DSP/BRAM usage within the available reconfigurable fabric. A significant amount of BRAMs is needed for data and phase factor storage when the FFTs are computed with a high amount of points.
- the adoption of the zoom-in approach allows the total memory requirements per channel, to be kept at reasonable levels.
- the synchronous reset and start bit signals are obtained from a control register, allowing for an operation fully controlled by the PS.
- an additional squaring block is introduced at FIFO output.
- the communication between the several logic blocks on the reconfigurable fabric and the ARM GPP is ensured by specifically defined control and data channels.
- a set of control registers are made accessible to the programs executing on the ARM, providing convenient software reset and start flags triggering functionalities. These registers are placed on an AXI bus, mastered by the CPU through an AXI-GP interface port.
- An AXI-HP port is used to transfer the FFT output frames directly to the main memory, where they can be subsequently accessed by the CPU.
- an AXI Burst Master Controller is used to provide these co-processing channels 144 with a direct memory access. This solution provides a data-transfer performance equivalent to a conventional DMA engine, but reduces the hardware requirements by approximately 7 times.
- the architecture in FIG. 3 is expanded to incorporate as many co-processing blocks as the number of channels, e.g., blocks 108/acquisition channels 116 of the biochip 102 (dashed box labeled Co-processing channel 144 in the lower part of Fig. 3 ).
- the co-processing blocks are then connected to the AXI-HP port, which provides a communication channel between the co-processors and the main memory capable of supporting data transfers of up to 1.2 GByte/s (estimate). This is well above the 72 KBit/s bit-rate of the ADC in certain biosensor applications.
- two different interfaces are supported by the biosensing platform: i) an SPI-based interconnection with each channel ADC and with stimuli generator, ii) the interface with the human operator.
- the connections to the biosensing ADC and with the stimuli generator can be accomplished by an open-source core implementation of the SPI protocol, available at OpenCores.org. The selection of this particular core takes into account its particular characteristics in terms of versatility and resource efficiency.
- the interface 106 e.g., graphical user interface
- the interface 106 is accomplished by using an HDMI graphics port and an USB keyboard, USB touch-screen panel or other suitable interface.
- an HDMI interface already exists in the Zynq platform, the controller has to be implemented on the reconfigurable fabric.
- the controller module was obtained from a reference design targeting an HDMI output interface, along with the physical and data streaming blocks required to handle DMA video streams. These DMA engines are connected to two separate AXI-HP ports, enabling them to directly gather video data from the main memory.
- the HDMI control logic (also on the PL side), is connected to an AXI-GP interface and is mapped into the memory-space of the OS, allowing the usage of existing software device drivers.
- a method 400 of operating a biological detection system having a biochip is characterized.
- the method 400 may be used, for instance, to calibrate drive circuitry signals for the best signal to noise ratio.
- the method 400 is implemented by setting at 402, drive circuitry to generate a set of drive signals and reading at 404, sensor data from a biochip.
- the method 400 is also characterized by performing at 406, digital processing, with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip.
- the method 400 is further characterized by using the result of the digital processing with a low frequency resolution to define at 408, a preliminary indication of a target frequency to measure a sensor data output from the signal acquisition channel, and by identifying at 410, a range or band of target frequencies. Yet further, the method is characterized by performing at 412, filtering and signal demodulation to displace a frequency (band) of interest from the identified range of target frequencies down to a base band and by designating at 414, a bit and frequency resolution from the base band that maximizes the signal to noise ratio of the data read from the signal acquisition channel.
- the method may be further characterized by performing demodulation where sensor data extracted from the biochip results from using modulation techniques.
- performing digital processing is further characterized by performing a fast Fourier transform with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip.
- the method may also be further characterized by performing complex correction algorithms including at least one of a: Kalman filter, drift compensation, and a temperature controller, on the result from the Fourier transform.
- bit-streams with smaller FFT cores were generated and the number of channels was maximized by balancing the usage of DSPs, BRAMs and LUTs.
- Table 2 presents the resources used by a single co-processing channel 144, considering FFTs with increasing number of points, and the maximum number of co-processing channels 144 supported on the PL. It should be noticed that for the 1k-, 2k- and 4k-FFT the maximum number of channels is constrained by the number of available DSP48 elements; for larger FFTs the limitation lies in the BRAMs. Also, in the 32k-FFT, to support 2 channels, some block RAMs were implemented using LUTs.
- the AXI-HP allows an estimated maximum transfer rate of 1.2 GByte/s, while, for a data width of 32 bits, the FFT core outputs a bit-rate of at most 128 Mbit/s.
- the AXI-HP bandwidth does not limit the system operation, supporting multiple acquisition channels. Considering that the developed biochip features 4 sensor blocks (1024 sensors), the bit rate could still be increased if faster ADCs become available.
- systems are provided to deliver the computational power required by advanced biochip technology, and to allow simultaneous acquisition and processing at high sampling rates, through a high-performance and configurable acquisition and processing architecture.
- a biosensing platform which uses multiple dedicated co-processors performing Fast Fourier Transform (FFT)-based signal filtering, as well as other complex correction and modulation/multiplexing algorithms.
- FFT Fast Fourier Transform
- An additional dedicated processor is used to synthesize the stimulus signals for the sensors, and a central processor is provided to control the whole system.
- a fully autonomous embedded system is provided, which enables standalone operation of the platform, thus providing increased flexibility and usability.
- the disclosure also provides a biological detection system (100), characterized by:
- the biochip is further characterized by:
- the set of acquisition interfaces is further characterized by:
- the analog to digital converter has a sampling frequency of at least 1 kHz and at least 12-bit resolution.
- each co-processor is characterized by:
- system is further characterized by:
- the drive circuitry is further characterized by:
- the system is characterized by:
- the drive circuitry is characterized by:
- an acquisition interface is configured to measure sensor data from the corresponding signal acquisition channel at a specific binding frequency or frequency band that is determined by the modulation induced by the magnetic field, the bias current, the sensor's response to these AC signals, and the system's noise and cross-talk.
- the system is characterized by:
- the disclosure also provides a method of operating a biological detection system having a biochip, characterized by:
- the method is further characterized by:
- performing digital processing is further characterized by:
- the method is further characterized by:
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- General Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Medicinal Chemistry (AREA)
- Human Computer Interaction (AREA)
- Hematology (AREA)
- Molecular Biology (AREA)
- Urology & Nephrology (AREA)
- Food Science & Technology (AREA)
- Electrochemistry (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Apparatus Associated With Microorganisms And Enzymes (AREA)
- Image Processing (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
Description
- The present disclosure relates in general to biological analysis, and in particular to scalable, multi-channel high performance biosensing platforms, which are capable of high throughput biological analysis.
- Conventional biological analysis methods are being gradually replaced by sophisticated and compact biochip-based point-of-care devices. These biochip-based point-of-care devices perform common blood or glucose analysis at a fraction of the cost of conventional biological analysis methods. Also, biochip-based point-of-care devices typically require a much shorter analysis time compared to conventional biological analysis methods. Given the high potential of these biochip-based point-of-care technologies, research in biodetection methods and in biochips themselves, has significantly grown in the past decade.
- From continued research in biodetection methods, new and more efficient biological analysis methods have become possible, providing a valuable tool for several applications, namely, but not limited to: clinical diagnostics, forensics, veterinarian, food industry, agriculture, and drug screening.
- According to aspects of the present disclosure, a biological detection system is provided. The biological detection system is characterized by a biochip and a biosensing platform. The biochip has at least two blocks of sensors, each block of sensors having a signal acquisition channel that is independently accessible such that each block of sensors can be read in parallel. The biosensing platform has drive circuitry to drive each block of sensors of the biochip. In an exemplary implementation, the same or different driving signals can be applied to each block of sensors of the biochip. Also, the biosensing platform further has a set of acquisition interfaces arranged such that a select one acquisition interface is coupled to a corresponding one signal acquisition channel of the biochip. The biosensing platform also has a set of co-processors such that a select co-processor is uniquely associated with a corresponding one of the acquisition interfaces. Still further, the biosensing platform has a central controller that controls the drive circuitry. The central controller also interfaces with the set of acquisition interfaces. Still further, the central controller interfaces with the set of co-processors to read data from the biochip. Moreover, the biosensing platform may provide a direct user interface, a remote interface or combination thereof.
- According to further aspects of the present disclosure, a method of operating a biological detection system having a biochip, is provided. The method may be used, for instance, to calibrate drive circuitry to achieve a desired signal to noise ratio of signal data read from the biochip. The method is characterized by setting drive circuitry to generate a set of drive signals and by reading sensor data from the biochip. The method is also characterized by performing digital processing with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip. The method is further characterized by using the result of the digital processing with a low frequency resolution to define a preliminary indication of a target frequency to measure a sensor data output from the signal acquisition channel, and by identifying a range or band of target frequencies. Yet further, the method is characterized by performing filtering and signal demodulation to displace a frequency (band) of interest from the identified range of target frequencies down to a base band and by designating a bit and frequency resolution from the base band that maximizes the signal to noise ratio of the data read from the signal acquisition channel.
- The method may be further characterized by using a modulation technique to drive the biochip and correspondingly, performing demodulation to extract sensor data from the biochip. In another exemplary implementation, performing digital processing is further characterized by performing a fast Fourier transform with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip. In yet another exemplary implementation, digital signal processing may be implemented for performing one or more complex correction algorithms, which may include a Kalman filter, drift compensation, and a temperature controller, on the result from the Fourier transform.
-
-
FIG. 1 is a schematic representation of a system for biological analysis that integrates a biochip and an embedded biosensing platform, according to aspects of the present disclosure; -
FIG. 2A is an exploded view schematic representation of a biochip of the system for biological analysis illustrated inFIG. 1 ; -
FIG. 2B is an exploded view schematic representation of a biosensing platform of the system for biological analysis illustrated inFIG. 1 ; -
FIG. 2C is an exploded view schematic representation of an interface of the system for biological analysis illustrated inFIG. 1 ; -
FIG. 3 is a block diagram of a biosensing platform, according to aspects of the present disclosure herein; and -
FIG. 4 is a method of operating a biological detection system having a biochip and a biosensing platform, according to aspects of the present disclosure. - According to various aspects of the present disclosure, systems, methods, and computer implemented processes are utilized to implement scalable, multi-channel high performance biosensing platforms, which are capable of high throughput biological analysis. For instance, aspects of the present disclosure find use in applications such as gene expression profiling, where thousands of biosensors analyze a given sample in parallel. In this regard, the biosensing platform described herein, integrates the complexity and computational needs of the electronic sensing system, into portable and standalone embedded systems with adequate computational capabilities for such high throughput analysis.
- According to aspects of the present disclosure, systems are provided to deliver the computational power required by advanced biochip technology, and to allow simultaneous acquisition and processing at high sampling rates, through a high-performance and configurable acquisition and processing architecture. For instance, as will be described in greater detail herein, a biosensing platform is provided, which uses multiple dedicated co-processors performing Fast Fourier Transform (FFT)-based signal filtering, as well as other complex correction algorithms. An additional dedicated processor is used to synthesize the stimulus signals for the sensors, and a central processor is provided to control the whole system. With this approach, a fully autonomous embedded system is provided, which enables standalone operation of the platform, thus providing increased flexibility and usability.
- Referring now to the drawings, and in particular, to
FIG. 1 , abiological detection system 100 is illustrated in schematic form, which is capable of biological analysis according to aspects of the present disclosure. In general, thebiological detection system 100 includes abiochip 102 and abiosensing platform 104. - The
biological detection system 100 also integrates with aninterface 106 to provide a complete lab-on-chip solution. Theinterface 106 may include for instance, local user interface peripherals (such as a keyboard, mouse, display, etc.), a remote interface, or a combination thereof, which can exchange data with electronics of thebiosensing platform 104. - Referring generally to
FIG. 1 andFIG. 2A , the biochip 102 (e.g., a magnetoresistive biochip) includes one or more "blocks" 108 of sensors. For instance, as illustrated, there are n blocks of sensors where n is any integer greater than zero. In this manner, particular blocks are referenced as 108-1, 108-2, ..., 108-n, etc. Eachblock 108 of sensors has asignal acquisition channel 110 that may be independently accessible such that eachblock 108 of sensors can be read in parallel, as will be described in greater detail herein. The biochip sensors (e.g., theblocks 108 of sensors) achieve a high sensitivity, high Signal-to-Noise Ratio (SNR) and high dynamic range when used in conjunction with thebiosensing platform 104 described herein. - As best illustrated to the left of the
biochip 102, an exploded view (of block 108-1 by way of example) illustrates that eachblock 108 includes a plurality of sensors. Each of the plurality of sensors can be of a various type (e.g. magnetic, temperature). By way of illustration, and not by way of limitation, there are 256 sensors in eachblock 108, arranged in an array of 16 rows and 16 columns. According to aspects of the present disclosure, more than 256 sensors are required for many important applications (e.g., gene expression profiling). To address this requirement, theillustrative biochip 102 combines CMOS with thin film fabrication technologies to provide the required number of sensors on a small biochip package. Thus, for instance, abiochip 102 with over 1,000 sensors is realized by implementing n=4blocks 108, each with 256 sensors. However, in practice, eachblock 108 can have any number of sensors in its array. For instance, eachblock 108 can include 16 sensors or more in certain illustrative implementations. - In certain illustrative implementations, each
block 108 of thebiochip 102 may have a single acquisition channel. Each acquisition channel (or block of sensors themselves) can include integrated on-chip signal conditioning circuitry, and necessary readout circuitry to address and access the individual sensors of acorresponding block 108. For instance, by using a sensor reading technique, such as Time-Division Multiplexing (TDM), a control device is able to address each of the biosensors (256 biosensors in the above-example) through that single channel. For instance, each sensor can be read out individually using time-multiplexed reading techniques, such as by using a first multiplexer ("row multiplexer") to select a desired row, and a second multiplexer ("column multiplexer") to select a desired column. The sensor that is read out through the signal acquisition channel is thus the sensor at the intersection of the row selected by the row multiplexer and the column selected by the column multiplexer. In an illustrative implementation, thebiochip 102 is characterized by each block of sensors having an array of at least 16 sensors and a corresponding time division multiplexer to read out individual sensors from a corresponding block. Moreover, in practical implementations, since there aremultiple blocks 108 of sensors, thebiochip 102 can provide a multichannel capability that includes up to, and/or in excess of 1000 sensors in certain exemplary implementations. - Referring now to
FIG. 1 andFIG. 2B , thebiosensing platform 104 includes in general,drive circuitry 112 that generates the necessary drive signal(s) for thecorresponding biochip 102. More particularly, thebiosensing platform 104 includesdrive circuitry 112 to drive eachblock 108 of sensors of thebiochip 102. In this regard, the manner in which the drive signals from thebiosensing platform 104 couple to thebiochip 102 will depend upon the particular implementation of thebiochip 102. However, by way of example, the drive circuitry may include circuitry to generate bias currents and oscillating magnetic fields that are applied to the sensors of thebiochip 102. - For instance, the
drive circuitry 112 can include a single circuit with multiple outputs. Alternatively, the drive circuitry can include multiple independent circuits, e.g., an independent circuit for eachblock 108 of thecorresponding biochip 102. Still further, the one or more circuits that make up thedrive circuitry 112 can be software-based, hardware based, or a combination of the above. In this regard, drive signals can be fully or partially programmable, controlled by set points, parameters or other program variables, made adjustable by control signals, e.g., programmed by control voltages or other control signals, etc. - As illustrated, the
drive circuitry 112 is controlled by acentral processor 114. Thus for example, in an illustrative implementation, thedrive circuitry 112 is controlled by thecentral processor 114 to have an adjustable frequency. Moreover, thecentral processor 122 can control the drive circuitry to generate arbitrary signal shapes and amplitudes. For instance, thedrive circuitry 112 can utilize Direct Digital Synthesizers (DDSs) that can be combined with arbitrary signal generators deployed in a reconfigurable fabric, which are controlled by thecentral processor 114, to provide the correct signals to thecorresponding biochip 102. - The
biosensing platform 104 also includes multiple (e.g., a set of) acquisition interfaces 116. As illustrated, thebiosensing platform 104 includes n acquisition interfaces 116 where n is any integer greater than zero. In the illustrated example, a select oneacquisition interface 116 is coupled to a corresponding onesignal acquisition channel 110 of eachblock 108 of thebiochip 102. Thus, where thebiochip 102 has 4blocks 108, each block having a single,signal acquisition channel 110, thecorresponding biosensing platform 104 includes 4 acquisition interfaces 116, eachacquisition interface 116 connected to a corresponding unique one of the biochipsignal acquisition channels 110. - In the illustrative schematic representation, each
acquisition interface 116 includes aconverter 118, e.g., an analog to digital converter, which receives the signal from its correspondingsignal acquisition channel 110. The analog to digital converter converts the sensor signal into a digital representation that can be further processed, as will be described in greater detail herein. As an illustrative example, anacquisition interface 116 is configured to measure sensor data from the correspondingsignal acquisition channel 110 at a specific binding frequency or frequency band that is determined by the modulation induced by the magnetic field, the bias current, the sensor's response to these AC signals, and the system's noise and cross-talk. For instance, each analog todigital converter 118 may have a sampling frequency of at least 1 kHz and at least 12-bit resolution to accomplish the desired sampling. - The output of the analog to
digital converter 118 couples to acontroller 120, e.g., a serial peripheral interface controller. Eachacquisition interface 116 may also include signal conditioning circuitry as necessary to interface with the analog signals from thebiosensor 102. Thecontrollers 120 are each controlled by thecentral processor 114. - Moreover, in an exemplary implementation, each
controller 120 couples to acorresponding co-processor 122. In this manner, eachcontroller 120 conveys information from its corresponding analog todigital converter 118 to itscorresponding co-processor 122. Thus, the illustratedbiosensing platform 104 includes a set of co-processors 122 (e.g., n co-processors where n is any integer greater than zero) such that aselect co-processor 122 is uniquely associated with a corresponding one of the acquisition interfaces 116. In alternative implementations, there need not be a one-to-one correspondence betweenacquisition interfaces 116 andco-processors 122. For instance, there may be asingle co-processor 122 for all channels (acquisition interfaces 116). Also, asingle co-processor 122 can be utilized to service a subset of acquisition interfaces 116. - Each
dedicated co-processor 122 performs digital processing, e.g., Fast Fourier Transform (FFT)-based signal filtering on the data from the corresponding acquisition interface. Each co-processor 122 can also/alternatively perform other complex correction algorithms, e.g., a complex correction algorithm including a Kalman filter. Yet further, each co-processor 122 can be used to implement a temperature controller, a complex modulator and demodulator, etc. Theco-processors 122 also communicate with thecentral processor 114, which controls thebiosensing platform 104. - In this manner, to read data from the biochip, the
central processor 114 of thebiosensing platform 104 controls thedrive circuitry 112. Thecentral processor 114 also interfaces with the set of acquisition interfaces 116 (e.g., via the controllers 120). Yet further, thecentral processor 114 interfaces with the set ofco-processors 122. Also, thecentral processor 114, adedicated co-processor 122 or other processor (not shown) can be utilized to synthesize the stimulus signals for the data read from the sensors of thebiochip 102. - The illustrated
biosensing platform 104 further includes amain memory 124. The main memory can couple to thecentral processor 114 and/or to one or more of the co-processors 122. For instance, thecentral processor 114 and theco-processors 122 can share themain memory 124. In this regard,main memory 124 may be used to store programming, control information, calibration information, data, etc. As such, in certain applications, a computer processing device (e.g., personal computer or server computer) need not be connected to thebiosensing platform 104 in order for the system to be operational. This allows, for instance, flexible standalone use. At a suitable time, thebiosensing platform 104 can be coupled to the computer processing device, e.g., to download collected signal data, etc. - In this regard, the
biosensing platform 104 includes electronic circuitry for addressing thebiochip 102, for driving control signals and for conditioning and acquiring the biochip signals. The acquired signals are processed by a set ofco-processors 122 and are transmitted to thecentral processor 114, e.g., which executes software that functions as a digital analyzer where, through theinterface 106, e.g., a conventional graphical user-interface, the experiment is controlled by the operator or analyst. Thebiosensing platform 104 provides multiplesignal acquisition channels 110 operating at a high sampling rate, thus producing a high volume of data. As such, thebiosensing platform 104 also includes the necessary computation power that is required to handle the volume of data. - Referring to
FIG. 1 andFIG. 2C , theinterface 106 may comprise a monitor, keyboard, mouse or other computer peripherals, a graphical user interface, network interface, etc. In an illustrative implementation, the biological detection system includes a biosensing platform implemented in two components, including a programmable logic component and a processing system component, wherein the drive circuitry 112 (e.g., drive signal synthesizer), theco-processors 122 and the acquisition interfaces 116 are implemented in the programmable logic component, and thecentral processor 114 and themain memory 124 are implemented in the processing system. - The
central processor 114 may further communicate over a network, which provides communications links between various processing devices, and may be supported by networking components that interconnect the processing devices, including for example, routers, hubs, firewalls, network interfaces, wired or wireless communications links and corresponding interconnections, cellular stations and corresponding cellular conversion technologies, e.g., to convert between cellular and tcp/ip, etc. Moreover, the network(s) may comprise connections using one or more intranets, extranets, local area networks (LAN), wide area networks (WAN), wireless networks (WIFI), the Internet, including the world wide web, cellular and/or other arrangements for enabling communication between the processing devices, in either real time or otherwise, e.g., via time shifting, batch processing, etc. - With reference to
FIG. 1 , andFIGS. 2A-2C generally, a laboratory assay begins by labeling biological targets under analysis with Magnetic Particles (MPs). Then, a bias current, iB, and an oscillating magnetic field signal, hB, are applied to the sensors of thebiochip 102, e.g., via thedrive circuitry 112 of thebiosensing platform 104. Upon biomolecular recognition and washing, the sensors detect the fringe fields created by the MPs that have been captured by biological probes placed above the sensors. The biological information is then extracted by measuring the variations of the sensor voltage signal, vbind . - To reduce the system cost and complexity, the sensors of each
block 108 are measured using Time-Division Multiplexing (other modulation and multiplexing techniques can alternatively be applied). Since reaching a low noise level typically requires a high number of samples per sensor, biochip reading times can be slow, e.g., about 4 minutes. Such biomolecular recognition at a low speed, may be acceptable for certain biochips. However, such slow recognition does not scale when the number of sensors grows to 1000 or more. To circumvent issues of slow speeds, thebiochip 102 is organized in groups of sensors, e.g., 256 sensor elements perblock 108, and each sensor is time multiplexed within its corresponding block, as noted above. With this approach, the total reading time is constrained to the number of sensors within each group, as processing is performed in parallel. Thus biochip size can be scaled without increasing the overall reading time. Nevertheless, this also imposes significant hardware resources in the platform electronics. Eachgroup 108 of thebiochip 102 is serviced by a dedicatedacquisition channel interface 116 of thebiosensing platform 104, e.g., comprising signal conditioning circuits and an Analog-to-Digital Converter (ADC) 118. Furthermore, the data from each channel is processed in parallel, to allow the observation of the assay evolution in real-time. - Additional characteristics of the biochip must also be considered. According to an exemplary implementation, to avoid the effect of low frequency noise and maximize the attained SNR, the applied signals from the
drive circuitry 112, e.g., applied signals iB and hB in the FIGURES, include an AC component. Consequently, vbind is measured at a specific binding frequency, f bind, which is determined by the modulation induced by the magnetic field hB {f2 }, the bias current iB {f1 }, the sensor's response to these AC signals, and the system's noise and cross-talk. At each sensor block multiple modulation frequencies can be applied in both magnetic field and bias current. - Due to the complexity of the biochip technology and of the circuitry that drives and measures the sensors, it may not be easy to determine a good a-priori estimate of the frequency at which the SNR is maximum. To optimize the biological sensitivity, fbind should be adapted in real-time, by changing the frequencies of iB and/or hB. However, this introduces the added difficulty of determining the signal amplitude at an unknown frequency. According to illustrative aspects of the disclosure herein, a flexible and tractable method is to use an FFT of the signal acquired at each of the 1000+ sensors, to compute the signal amplitude vbind at the frequency fbind . Although it ensures a good adaptability, this approach may require a sampling frequency of at least 50 kHz and a frequency resolution as low as 1 Hz, for a 24-bit resolution in each sensor. That said, aspects of the present invention can function suitably, for instance, with a sampling frequency of at least 1 kHz and at least a 12-bit resolution. Regardless, this imposes large memory and hardware requirements, constraining the scalability of the system. Additionally, other complex correction algorithms can be implemented by each co-processor 122, namely but not limited to: drift correction, temperature controller, Kalman filters, complex modulator and/or demodulator (e.g. spread spectrum).
- The computational performance required by the
biosensing platform 104 to ensure the resolution and reliability of bio-recognition assays can be achieved by optimizing the processors' architecture and by increasing the exploited parallelism with the usage of multicore structures. Nonetheless, it would be insufficient to exclusively address the performance limitations, since the bottleneck may also arise in the communication channels, such as the one connecting the acquisition circuits to the processing unit. Furthermore, due to the significant complexity of the underlying algorithms, the processing unit must be able to store and manipulate large datasets, making it clear that a high memory bandwidth and capacity are necessary to take full advantage of the enhanced processing capabilities. - As described more fully herein, the architecture to fulfill the requirements of the high throughput biosensing platform is depicted in
FIG. 1 . In general, when implementing thebiosensing platform 104, a high-speed serial interface may be selected in order to provide a fast communication between theADCs 118 and the set of dedicated co-processors 122 (e.g., cores) where the most computational intensive tasks are handled. All components, including the generator of the stimuli signals, are configured by acentral processor 114, e.g., also referred to herein as a General Purpose Processor (GPP) which, aside from acting as the main system controller, can also perform auxiliary computations on the acquired data and user interface tasks as well as communication tasks. Thecentral processor 114 shares itsmain memory 124 with the co-processors 122. A user interface, controlled through thecentral processor 114, is preferable performed throughinterface 106 which includes a set of I/O devices, such as the input peripherals used to control the experiment and a display on which the results are presented to the user. A network interface, controlled through thecentral processor 114, is also included, allowing the processed data to be transferred to an external database, easily available to other users. - The
central processor 114, e.g., the GPP, is a suitable device to configure the several system components, and thus thecentral processor 114 plays a key role on easing the system integration. In certain illustrative implementations, Operating System (OS) support is necessary, in order to benefit from the available software packages that provide complex system management functionalities and high-level interfaces to the standard I/O and memory peripherals. A processor-memory interconnection subsystem (subsystem that interconnects thecentral processor 114 to the main memory 124) can be equipped with DMA channels (or equivalent), allowing thecentral processor 114 to communicate with streaming peripherals within a shared memory space. - Each acquisition channel is linked to one
biochip sensor group 108 and directly feeds onespecialized co-processor 122 with the data sampled by theADC 118. In an exemplary implementation, since eachADC 118 can operate at a maximum rate of 96 Mbit/s, the absence of a high-speed data channel would seriously compromise the overall throughput. In the exemplary implementation, standard Low-Voltage Differential Signaling (LVDS) was chosen as a cost-effective and noise tolerant solution to achieve high transmission rates. On top of this physical layer, a Serial Peripheral Interface (SPI) 118 bus protocol is utilized, to provide configuration and addressing capabilities based on an SPI master/slave controller. - The
specialized co-processors 122 are responsible for measuring the amplitude of the sensors binding signal, vbind , from thebiosensor 102, at a specific frequency. This signal is estimated using a zoom-in FFT approach together with other complex processing algorithms like Kalman filters that improve the overall data accuracy. In a first step, an FFT is performed with a low frequency resolution, and by using a subset of sensors in the same channel. This provides a coarse and preliminary identification of the best fbind frequency to measure the vbind , while keeping the hardware resources and power consumption at tolerable levels. After identifying the fbind range, band-pass filtering and signal demodulation are performed, to displace the frequency of interest down to base band. This allows the computation of multiple FFTs (one per channel) with a higher bit and frequency resolution, while still confining the required resources. Also, to increase the signal to noise ratio and achieve system scalability, the zoom-in procedure can be performed as many times as required. Complex demodulators (e.g. spread spectrum) can also be implemented in the co-processors to achieve a higher multiplexing capability. - Due to the involved computational complexity, a reconfigurable field programmable gate array (FPGA) device is regarded as a suitable technology to accommodate the multi-core infrastructure of specialized FFT co-processors, filters and
demodulators 122. An FPGA combines the required high performance level, while still conferring the flexibility to change or update the adopted processing algorithms. In particular, multiple configurations can be stored in memory for each zoom-in level. In run-time, the user starts by observing the coarse signal spectrum of the initial subset of sensors. Then, the most favorable demodulation frequency and spectrum band is selected. This will trigger the FPGA reconfiguration and the deployment of the specialized co-processors in hardware, configured with the target demodulation frequency. At this point, the user can observe the binding signal for each sensor. - A set of I/
O peripherals 106 are utilized to make the biosensing platform 104 a standalone analysis and diagnosis system. These modules not only allow the user to take full control of the experiment, but also offer the visualization means for clinical interpretation purposes. The standard set of computer peripherals (i.e. USB keyboard / touch screen panel and HDMI monitor) is typically utilized as interfaces to thebiosensing platform 104 to enhance the usability experience. - Referring now to
FIG. 3 , a portion of thebiological detection system 100 ofFIG. 1 is illustrated. In particular,FIG. 3 illustrates a partial view of thebiosensing platform 104. For instance, only a singlesignal acquisition interface 116 is illustrated for convenience of illustration and discussion. Also, thebiochip 102 and theinterface devices 106 have been omitted for convenience of discussion. - Reference AC signals must be supplied by the
drive generator 112 to the biosensors of thebiochip 102, to produce a measurable response (e.g., iB and hB discussed above). The amplitude, shape and frequency ranges of these stimuli are partially constrained by the considered biochip technology. However, the noise level that the measurements are exposed to is inversely proportional to the frequency of these signals. Thus the accuracy of the results is strongly constrained by this parameter. As a consequence, a variable and configurable frequency generator component of thedrive circuitry 112, which can be adjusted and parametrized by the central processor 114 (e.g., GPP), ensures the adaptability of thebiosensing platform 104 to multiple generations of thebiochip 102. The required signal generators can be implemented by using Direct Digital Synthesizers (DDSs) deployed in a reconfigurable fabric, which can be combined with additional hardware in the reconfigurable fabric to achieve complex modulation like spread spectrum, and controlled by thecentral processor 114. - In a working exemplary embodiment, a Xilinx Zynq 7000 Extensible Processing Platform was utilized to build the
biosensing platform 104 as an embedded system. This particular System on a Chip (SoC) includes a state of the art reconfigurable fabric and it offers most of the features required for the system described more fully with regard toFIGS. 1-3 . - The Zynq SoC integrates two independent modules in the same chip: a Processing System (PS), featuring a
processor 132, e.g., a dual-core ARM Cortex-A9 processor, and Programmable Logic (PL), consisting of reconfigurable fabric compatible with Xilinx 7-Series FPGAs. In addition to the ARM processors and the associated DDR-memory controller, the PS also integrates a collection of industry-standard communication interfaces 134 (Gigabit Ethernet, USB, I2C and SPI). At boot time, the PS configures the PL by means of the Processor Configuration Access Port (PCAP) and initializes the system clocks. Through the PCAP, the ARM processor is able to exploit the software-driven dynamic partial reconfiguration capabilities of the reconfigurable fabric, thus promoting an adapted and optimized usage of the reconfigurable part of this SoC. - The PS-PL interconnection can be achieved by using an interface, e.g., the Advanced eXtensible Interface (AXI) bus protocol, through any of the provided connectivity ports, namely, General-Purpose (AXI-GP), High-Performance (AXI-HP), and Accelerator Coherency Port (AXI-ACP). For instance, as illustrated, two master and two slave AXI-
GP interfaces 136 connect the PL to a PScentral interconnect 138, allowing for low throughput data and control transfers. When high throughput is required, aseparate interface 140, e.g., an AXI-HP interface, connects directly to themain memory 142. In thisinterface 140, read and write operations are mediated by an asynchronous 1 kB FIFO with a configurable data width of 32 or 64 bits. The AXI-ACP interface connects the PL directly to the Snoop Control Unit (SCU), to ensure data coherency at the L2 shared cache of the ARM processors. All these ports can be individually enabled and configured. On the PL side, these interfaces are implemented using the existing reconfigurable logic, thus having an associated cost in terms of hardware resources. - While the ARM processor on the PS is used to control the platform, the reconfigurable fabric on the PL is used to implement the dedicated co-processors and to support the interface with the biochip. The base software layer running in the PS is implemented in an exemplary implementation, using the Linaro Ubuntu GNU/Linux OS, which provides optimized tools and software packages for ARM SoCs. By adopting this OS, the PS provides a rich interface to the user by using the existing set of computer peripherals.
-
FIG. 3 depicts the block diagram of the implemented architecture, considering a single-channel 116. It corresponds to the initial zoom-in phase, where an FFT is computed over a small subset of sensors from a single acquisition channel to determine the binding frequency, fbind that maximizes the signal to noise ratio. In the multi-channel implementation, where parallel channels read biological information of 1000+ sensors, the Co-processor channel 1X inFig. 3 is replicated as necessary, according to theblocks 108/acquisition channels 116 of thebiochip 102. - Several AXI interfaces connect the reconfigurable fabric to the ARM processor and to the associated main memory, either for control or for high throughput data transfers. By following the design strategy commonly adopted in DMA-based peripherals, each devised co-processor provides a slave interface for configuration purposes, along with a master controller that writes in the shared memory bus. Data consistency between the dedicated co-processors and the GPP is achieved by using a contiguous region in the main memory, marked as non-cacheable and excluded from the system's memory management. The OS is able to read this memory region by remapping the region into the kernel virtual address space, thus allowing it to be easily accessed by a custom device driver.
- In an illustrative implementation, the FFT co-processors (e.g., co-processor 122) in the PL are based on the Xilinx FFT IP core v7.1, and adopt a pipeline operation to produce an output per clock cycle in a stream-based processing scheme. The cores are configured to receive 24-bit fixed-point real inputs, although a higher precision is used inside the cores to decrease numerical error propagation between intermediate stages. Furthermore, embedded DSP blocks are used in the butterfly computations to allow for a better balance in the DSP/BRAM usage within the available reconfigurable fabric. A significant amount of BRAMs is needed for data and phase factor storage when the FFTs are computed with a high amount of points. However, the adoption of the zoom-in approach allows the total memory requirements per channel, to be kept at reasonable levels. In all zoom-in levels, the synchronous reset and start bit signals are obtained from a control register, allowing for an operation fully controlled by the PS. Also, since the considered signal processing algorithm requires the squaring of the FFT complex output, an additional squaring block is introduced at FIFO output.
- In an illustrative implementation, the communication between the several logic blocks on the reconfigurable fabric and the ARM GPP is ensured by specifically defined control and data channels. By following a memory-mapped I/O approach, a set of control registers are made accessible to the programs executing on the ARM, providing convenient software reset and start flags triggering functionalities. These registers are placed on an AXI bus, mastered by the CPU through an AXI-GP interface port. An AXI-HP port is used to transfer the FFT output frames directly to the main memory, where they can be subsequently accessed by the CPU. Since this shared memory region is contiguous and is not directly managed by the OS kernel (as explained above), an AXI Burst Master Controller is used to provide these
co-processing channels 144 with a direct memory access. This solution provides a data-transfer performance equivalent to a conventional DMA engine, but reduces the hardware requirements by approximately 7 times. - When multichannel processing is considered, the architecture in
FIG. 3 is expanded to incorporate as many co-processing blocks as the number of channels, e.g., blocks 108/acquisition channels 116 of the biochip 102 (dashed box labeledCo-processing channel 144 in the lower part ofFig. 3 ). The co-processing blocks are then connected to the AXI-HP port, which provides a communication channel between the co-processors and the main memory capable of supporting data transfers of up to 1.2 GByte/s (estimate). This is well above the 72 KBit/s bit-rate of the ADC in certain biosensor applications. Even considering that higher bit-rates are required for the disclosed biosensing platform, this bandwidth is enough to support multiple channels (thebiochips 102 comprise 4 channels), and still provides enough resources for future expansions. However, it should be noticed that a continuous data transmission must be assured to achieve a throughput of 1.2 GByte/s. To guarantee this requisite, a dual clock domain FIFO memory is used to implement a bridge between the FFT processing channels and the AXI-HP bus connected to the main memory. - In an illustrative implementation, two different interfaces are supported by the biosensing platform: i) an SPI-based interconnection with each channel ADC and with stimuli generator, ii) the interface with the human operator. The connections to the biosensing ADC and with the stimuli generator can be accomplished by an open-source core implementation of the SPI protocol, available at OpenCores.org. The selection of this particular core takes into account its particular characteristics in terms of versatility and resource efficiency.
- Referring generally to
FIG. 1 ,FIG. 2C , andFIG. 3 , in an illustrative implementation, theinterface 106, e.g., graphical user interface, is accomplished by using an HDMI graphics port and an USB keyboard, USB touch-screen panel or other suitable interface. Although an HDMI interface already exists in the Zynq platform, the controller has to be implemented on the reconfigurable fabric. The controller module was obtained from a reference design targeting an HDMI output interface, along with the physical and data streaming blocks required to handle DMA video streams. These DMA engines are connected to two separate AXI-HP ports, enabling them to directly gather video data from the main memory. The HDMI control logic (also on the PL side), is connected to an AXI-GP interface and is mapped into the memory-space of the OS, allowing the usage of existing software device drivers. - Referring to
FIG. 4 , according to further aspects of the present disclosure, amethod 400 of operating a biological detection system having a biochip, is characterized. Themethod 400 may be used, for instance, to calibrate drive circuitry signals for the best signal to noise ratio. Themethod 400 is implemented by setting at 402, drive circuitry to generate a set of drive signals and reading at 404, sensor data from a biochip. Themethod 400 is also characterized by performing at 406, digital processing, with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip. Themethod 400 is further characterized by using the result of the digital processing with a low frequency resolution to define at 408, a preliminary indication of a target frequency to measure a sensor data output from the signal acquisition channel, and by identifying at 410, a range or band of target frequencies. Yet further, the method is characterized by performing at 412, filtering and signal demodulation to displace a frequency (band) of interest from the identified range of target frequencies down to a base band and by designating at 414, a bit and frequency resolution from the base band that maximizes the signal to noise ratio of the data read from the signal acquisition channel. - In further aspects, the method may be further characterized by performing demodulation where sensor data extracted from the biochip results from using modulation techniques. In another exemplary implementation, performing digital processing is further characterized by performing a fast Fourier transform with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip. In this regard, the method may also be further characterized by performing complex correction algorithms including at least one of a: Kalman filter, drift compensation, and a temperature controller, on the result from the Fourier transform.
- To properly evaluate the scalable multichannel architecture, experimental results were obtained by using a ZedBoard featuring a Zynq XC7Z020, one of the smallest and less power demanding Zynq SoCs, and was synthesized and mapped using the Xilinx 14.3 development tools. The presented analysis starts by considering a single channel architecture implementation with the largest allowed FFT core, which is used to determine, with maximum accuracy, the biosensors binding frequency region. Experimental results are presented in Table 1 and show that the FFT-size (64k samples) is limited by the amount of available block RAMs in the PL. Furthermore, the results show that the resources required by the remaining structure corresponds to only 33.5% of the available LUTs, only 8 BRAMs and 19 DSPs. From this, it can be concluded that the majority of the PL hardware resources are made available to be used by the co-processing channel(s).
Table 1. System resources for a single channel 64K FFT. Component LUT (53200) BRAM (140) DSP48 (220) 64k FFT core 7157 (13.5%) 106 (75.7%) 103 (46.8%) Squaring module 149 (0.3%) 0 - 16 (7.2%) Drive signals (DDS) 174 (0.3%) 2 (1.4%) 3 (1.4%) AXI wrapper 572 (1.1%) 0 - 0 - SPI Master 131 (0.2%) 0 - 0 - Core system 13857 (26.0%) 6 (4.3%) 0 - Full system 25097 (47.2%) 114 (82.1%) 122 (55.4%) - To evaluate the scalability and adaptability of the architecture to the case when zoom-in is performed and multiple channels are required, bit-streams with smaller FFT cores were generated and the number of channels was maximized by balancing the usage of DSPs, BRAMs and LUTs. Table 2 presents the resources used by a
single co-processing channel 144, considering FFTs with increasing number of points, and the maximum number ofco-processing channels 144 supported on the PL. It should be noticed that for the 1k-, 2k- and 4k-FFT the maximum number of channels is constrained by the number of available DSP48 elements; for larger FFTs the limitation lies in the BRAMs. Also, in the 32k-FFT, to support 2 channels, some block RAMs were implemented using LUTs.Table 2. Co-processing channel hardware requirements. FFT size (#samples) Hardware resources (per channel) Maximum #channels LUT (53200) BRAM (140) DSP48 (220) 1024 3686 (7%) 8 (6%) 24 (11%) 8 2048 4375 (8%) 11 (8%) 31 (14%) 6 4096 4765 (9%) 14 (10%) 31 (14%) 6 16384 7201 (14%) 28 (20%) 40 (18%) 4 32768 7916 (15%) 55 (39%) 49 (22%) 2 65536 7254 (14%) 106 (76%) 103 (47%) 1 - The presented results suggest that it is possible to achieve a multichannel processing system composed of 8 FFT channels of 1024 points. Hence, with up to 256 biosensors per channel, a total of 2,048 biosensors can be processed and the sweep time is reduced by a factor of 8. Additional multiplexing can be achieved by using other complex modulation techniques like multi carriers or spread spectrum. This demonstrates the adaptability of the proposed system to multichannel processing, while keeping the required amount of hardware resources compatible with this entry level Zynq SoC. The FFT cores are capable of operating above 180 MHz, producing 1 output sample per clock cycle. Considering that the system must support an ADC sample rate of up to 4 Msps, which corresponds to the state of the art in high resolution converters, this suggests that the FFT cores do not impose any limitation to the signal analysis performance.
- On the other hand, the AXI-HP allows an estimated maximum transfer rate of 1.2 GByte/s, while, for a data width of 32 bits, the FFT core outputs a bit-rate of at most 128 Mbit/s. Hence, the AXI-HP bandwidth does not limit the system operation, supporting multiple acquisition channels. Considering that the developed biochip features 4 sensor blocks (1024 sensors), the bit rate could still be increased if faster ADCs become available.
- From these results, it can be concluded that the architecture herein, mapped on a Zynq SoC, provides the computational resources required to implement a fully embedded and high performance biosensing platform. Furthermore, the implemented prototype also demonstrated that a scalable and adaptable platform can be obtained with the existing reconfiguration capabilities, capable of adapting to biochips with higher sensor count and different computational needs.
- According to aspects of the present disclosure, systems are provided to deliver the computational power required by advanced biochip technology, and to allow simultaneous acquisition and processing at high sampling rates, through a high-performance and configurable acquisition and processing architecture. For instance, as described in greater detail herein, a biosensing platform is provided, which uses multiple dedicated co-processors performing Fast Fourier Transform (FFT)-based signal filtering, as well as other complex correction and modulation/multiplexing algorithms. An additional dedicated processor is used to synthesize the stimulus signals for the sensors, and a central processor is provided to control the whole system. With this approach, a fully autonomous embedded system is provided, which enables standalone operation of the platform, thus providing increased flexibility and usability.
- In this manner, new high density biochips are regarded as highly promising solutions for numerous applications, where high throughput analysis is required. However, existing electronic platforms are not able to acquire and process the higher amount of produced data. The scalable multi-channel and high-throughput architecture herein meets the needs to acquire and process this higher amount of data, e.g., even when implemented on a single Zynq SoC. For instance, even when using one of the smallest Zynq's, experimental results demonstrate that the system supports 8 times more sensors and a 1000+ times higher sampling frequency than the previous generation of the biosensing platform. By taking advantage of the reconfigurable characteristics of this SoC, results also suggest that the chosen platform allows to implement a scalable system supporting co-processors with configurable FFT sizes, enabling optimized and adjustable signal measurement for each biochip, without sacrificing efficiency in hardware resource usage, allowing the developed system to be expanded in order to deal with the requirements of future biochips.
- The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
- The corresponding structures, materials, acts, and equivalents of all means or step plus function elements in the claims below are intended to include any structure, material, or act for performing the function in combination with other claimed elements as specifically claimed. The description of the present invention has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the invention. Aspects of the invention were chosen and described in order to best explain the principles of the invention and the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
- The disclosure also provides a biological detection system (100), characterized by:
- a biochip (102) having at least two blocks (108) of sensors, each block of sensors having a signal acquisition channel (110) that is independently accessible such that each block of sensors can be read in parallel; and
- a biosensing platform (104) having:
- drive circuitry (112) to drive each block (108) of sensors of the biochip;
- a set of acquisition interfaces (116) arranged such that a select one acquisition interface (116) is coupled to a corresponding one signal acquisition channel (110);
- a set of co-processors (122) such that a select co-processor is uniquely associated with a corresponding one of the acquisition interfaces (116); and
- a central processor (114) that controls the drive circuitry (112), interfaces with the set of acquisition interfaces (116) and interfaces with the set of co-processors (122), to read data from the biochip.
- Optionally, the biochip is further characterized by:
- each block of sensors having an array of at least 16 sensors and a corresponding time division multiplexer to read out individual sensors from a corresponding block.
- Optionally, the set of acquisition interfaces is further characterized by:
- an analog to digital converter (118) and a controller (120), where the controller conveys information from the analog to digital converter to its corresponding co-processor.
- Optionally, the analog to digital converter has a sampling frequency of at least 1 kHz and at least 12-bit resolution.
- Optionally, each co-processor is characterized by:
- a configuration that performs at least one of:
- Fast Fourier Transform (FFT)-based signal filtering on the data from its corresponding acquisition interface;
- a complex correction algorithm including a Kalman filter,
- a temperature controller; and
- a complex modulator and demodulator.
- Optionally, the system is further characterized by:
- a main memory that is shared by the central processor and each of the co-processors.
- Optionally, the drive circuitry is further characterized by:
- circuitry to generate bias currents and oscillating magnetic fields that are applied to the sensors of the biochip.
- Optionally, the system is characterized by:
- the drive circuitry is coupled to the central processor such that a signal generator of the drive circuitry is controlled by the central processor to have an adjustable frequency and can also generate arbitrary signal shapes and amplitudes.
- Optionally, the drive circuitry is characterized by:
- Direct Digital Synthesizers (DDSs) that can be combined with arbitrary signal generators deployed in a reconfigurable fabric, which are controlled by the central processor.
- Optionally, an acquisition interface is configured to measure sensor data from the corresponding signal acquisition channel at a specific binding frequency or frequency band that is determined by the modulation induced by the magnetic field, the bias current, the sensor's response to these AC signals, and the system's noise and cross-talk.
- Optionally, the system is characterized by:
- the biosensing platform implemented in two components, including a programmable logic component and a processing system component, wherein the drive circutiry, the co-processors and the acquisition interfaces are implemented in the programmable logic component, and the central processor and the main memory are implemented in the processing system.
- The disclosure also provides a method of operating a biological detection system having a biochip, characterized by:
- setting drive circuitry to generate a set of drive signals;
- reading sensor data from a biochip;
- performing digital processing, with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip;
- using the result of the digital processing with a low frequency resolution to define a preliminary indication of a target frequency to measure a sensor data output from the signal acquisition channel;
- identifying a range or band of target frequencies;
- performing filtering and signal demodulation to displace a frequency (band) of interest from the identified range of target frequencies down to a base band; and
- designating a bit and frequency resolution from the base band that maximizes the signal to noise ratio of the data read from the signal acquisition channel.
- Optionally, the method is further characterized by:
- performing demodulation when sensor data extracted from the biochip results from using modulation techniques.
- Optionally, performing digital processing is further characterized by:
- performing a fast Fourier transform with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip.
- Optionally, the method is further characterized by:
- performing complex correction algorithms including at least one of a:
- Kalman filter, drift compensation, and a temperature controller, on the result from the Fourier transform.
Claims (4)
- A method of operating a biological detection system having a biochip, characterized by:setting drive circuitry to generate a set of drive signals;reading sensor data from a biochip;performing digital processing, with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip;using the result of the digital processing with a low frequency resolution to define a preliminary indication of a target frequency to measure a sensor data output from the signal acquisition channel;identifying a range or band of target frequencies;performing filtering and signal demodulation to displace a frequency (band) of interest from the identified range of target frequencies down to a base band; anddesignating a bit and frequency resolution from the base band that maximizes the signal to noise ratio of the data read from the signal acquisition channel.
- The method of claim 1, further characterized by:performing demodulation when sensor data extracted from the biochip results from using modulation techniques.
- The method of claim 1 or 2, wherein performing digital processing is further characterized by:performing a fast Fourier transform with a low frequency resolution on sensor data extracted from the biochip, using a subset of sensors that access a common signal acquisition channel of the biochip.
- The method of claim 3, further characterized by:performing complex correction algorithms including at least one of a:Kalman filter, drift compensation, and a temperature controller, on the result from the Fourier transform.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PT10713813 | 2013-08-30 | ||
EP14780768.9A EP2929341B1 (en) | 2013-08-30 | 2014-08-29 | Scalable and high throughput biosensing platform |
Related Parent Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP14780768.9A Division-Into EP2929341B1 (en) | 2013-08-30 | 2014-08-29 | Scalable and high throughput biosensing platform |
EP14780768.9A Division EP2929341B1 (en) | 2013-08-30 | 2014-08-29 | Scalable and high throughput biosensing platform |
Publications (3)
Publication Number | Publication Date |
---|---|
EP2988124A2 true EP2988124A2 (en) | 2016-02-24 |
EP2988124A3 EP2988124A3 (en) | 2016-06-29 |
EP2988124B1 EP2988124B1 (en) | 2017-07-19 |
Family
ID=51660434
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP14780768.9A Not-in-force EP2929341B1 (en) | 2013-08-30 | 2014-08-29 | Scalable and high throughput biosensing platform |
EP15184108.7A Not-in-force EP2988124B1 (en) | 2013-08-30 | 2014-08-29 | Scalable and high throughput biosensing platform |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP14780768.9A Not-in-force EP2929341B1 (en) | 2013-08-30 | 2014-08-29 | Scalable and high throughput biosensing platform |
Country Status (7)
Country | Link |
---|---|
US (2) | US20160161466A1 (en) |
EP (2) | EP2929341B1 (en) |
CN (1) | CN105849547A (en) |
DK (1) | DK2929341T3 (en) |
ES (1) | ES2610384T3 (en) |
PT (1) | PT2929341T (en) |
WO (1) | WO2015028642A2 (en) |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106681949B (en) * | 2016-12-29 | 2019-12-10 | 北京四方继保自动化股份有限公司 | Direct memory operation implementation method based on consistency acceleration interface |
CN107196695A (en) * | 2017-04-07 | 2017-09-22 | 西安电子科技大学 | Inter-satellite Links test system based on Zynq |
US11054243B2 (en) * | 2017-04-26 | 2021-07-06 | The Texas A&M University System | Electronic device for automatic calibration of swept-source optical coherence tomography systems |
CN108181003B (en) * | 2017-11-23 | 2023-11-14 | 深圳怡化电脑股份有限公司 | Zynq SOC-based infrared sensor control system and self-service terminal |
RU2763459C1 (en) * | 2018-05-23 | 2021-12-29 | Эксбайо Инк. | Integrated circuits for analysis of biological systems |
CN109491949B (en) * | 2018-11-27 | 2022-06-07 | 哈尔滨工业大学 | Zynq-based dynamic reconfigurable framework and method |
CN109965866B (en) * | 2019-01-30 | 2022-08-09 | 上海微瑞博生物科技有限公司 | Bioelectric signal acquisition module and signal acquisition card |
CN111159088A (en) * | 2019-11-29 | 2020-05-15 | 中国船舶重工集团公司第七0九研究所 | IIC bus communication method and system based on heterogeneous multi-core processor |
CN111182660B (en) * | 2019-12-31 | 2023-05-16 | 华东计算技术研究所(中国电子科技集团公司第三十二研究所) | Interactive wireless communication system based on general processor and ZYNQ processor |
CN111445380B (en) * | 2020-03-19 | 2021-01-19 | 熵智科技(深圳)有限公司 | Method and device for realizing real-time binocular stereo matching based on ZYNQ |
CN113079138A (en) * | 2021-03-23 | 2021-07-06 | 哈尔滨理工大学 | Online filtering system for biological information data |
CN113489594B (en) * | 2021-06-04 | 2023-12-19 | 北京中航双兴科技有限公司 | PCIE real-time network card based on FPGA module |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN2511972Y (en) * | 2001-11-28 | 2002-09-18 | 上海数康生物科技有限公司 | Capacitor biosensor for inspecting multiple biomolecules parallelly |
JP4399211B2 (en) * | 2002-12-21 | 2010-01-13 | 株式会社ハイニックスセミコンダクター | Biosensor |
CN2762136Y (en) * | 2005-01-18 | 2006-03-01 | 中国人民解放军军事医学科学院放射医学研究所 | Electrochemical array sensing chip |
AU2006212007A1 (en) * | 2005-02-11 | 2006-08-17 | The University Court Of The University Of Glasgow | Sensing device, apparatus and system, and method for operating the same |
KR100758285B1 (en) * | 2006-09-27 | 2007-09-12 | 한국전자통신연구원 | A bio sensor, method for manufacturing the same and a bio sensing apparatus with the same |
EP2095121A2 (en) * | 2006-12-18 | 2009-09-02 | Koninklijke Philips Electronics N.V. | Magnetic sensor device with suppression of spurious signal components |
CN101620221A (en) * | 2008-07-03 | 2010-01-06 | 汉述祖 | Bio-chip test device and test method thereof |
-
2014
- 2014-08-29 WO PCT/EP2014/068443 patent/WO2015028642A2/en active Application Filing
- 2014-08-29 CN CN201480060065.3A patent/CN105849547A/en active Pending
- 2014-08-29 EP EP14780768.9A patent/EP2929341B1/en not_active Not-in-force
- 2014-08-29 ES ES14780768.9T patent/ES2610384T3/en active Active
- 2014-08-29 PT PT147807689T patent/PT2929341T/en unknown
- 2014-08-29 DK DK14780768.9T patent/DK2929341T3/en active
- 2014-08-29 EP EP15184108.7A patent/EP2988124B1/en not_active Not-in-force
-
2016
- 2016-02-10 US US15/040,584 patent/US20160161466A1/en not_active Abandoned
-
2017
- 2017-10-18 US US15/786,837 patent/US20180038843A1/en not_active Abandoned
Non-Patent Citations (1)
Title |
---|
None |
Also Published As
Publication number | Publication date |
---|---|
CN105849547A (en) | 2016-08-10 |
ES2610384T3 (en) | 2017-04-27 |
DK2929341T3 (en) | 2017-01-30 |
EP2929341A2 (en) | 2015-10-14 |
US20180038843A1 (en) | 2018-02-08 |
PT2929341T (en) | 2016-12-30 |
US20160161466A1 (en) | 2016-06-09 |
WO2015028642A2 (en) | 2015-03-05 |
EP2929341B1 (en) | 2016-10-12 |
EP2988124A3 (en) | 2016-06-29 |
EP2988124B1 (en) | 2017-07-19 |
WO2015028642A3 (en) | 2015-06-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2988124B1 (en) | Scalable and high throughput biosensing platform | |
Hafizovic et al. | A CMOS-based microelectrode array for interaction with neuronal cultures | |
CN103690166B (en) | A kind of anti-imaging system of respiratory three-dimensional resistance based on PXI bus and formation method thereof | |
Wang et al. | Real-time, label-free monitoring of the cell cycle with a cellular impedance sensing chip | |
CN105891146A (en) | Intelligent terminal for spectrum detection and detection method thereof | |
CN110279416A (en) | A kind of portable impedance imaging system and its working method based on FPGA | |
Birk et al. | High-speed medical imaging in 3D ultrasound computer tomography | |
CN108333434A (en) | One kind dividing formula parallel organization capacitor array measuring circuit | |
CN105232050B (en) | A kind of noninvasive detection device of human body thickness of the tongue coating | |
CN203841697U (en) | Respiratory process three-dimensional electrical impedance tomography system based on PXI bus | |
US11346805B2 (en) | Handheld electrochemical sensing platform | |
Tiwari et al. | A Hybrid FPGA-GPU-based Hardware Accelerator for High Throughput 2D Electrical Impedance Tomography | |
Prasad et al. | Experimental setup of stretchable arid dry pad sensors for the signal acquisition fir filter design using Vedic approach | |
Rairigh et al. | Analysis of on-chip impedance spectroscopy methodologies for sensor arrays | |
Leitao et al. | Scalable and high throughput biosensing platform | |
CN116298198A (en) | Multi-parameter integrated soil sensor system and humidity fusion correction method | |
Li et al. | Reconfigurable bioimpedance emulation system for electrical impedance tomography system validation | |
Rymarczyk et al. | Multi frequency electrical tomography with re-configurable excitation waveforms | |
CN103808770B (en) | The early stage rapid and handy pick-up unit of cancerous tumor cell | |
Shaikh et al. | Development of Smart Fusion Technology Based Customizable System-on-Chip For Monitoring of Polyhouse Parameters | |
Wu et al. | Digital biomedical electrical impedance tomography based on FPGA | |
Pan et al. | Variational mode decomposition-based synchronous multi-frequency electrical impedance tomography | |
Al-Hashimi et al. | Design of a Reconfigurable, Modular and Multi-Channel Bioimpedance Spectroscopy System | |
Szumowski et al. | Hybrid tomographic device for acquisition of electrical tomography data | |
Sundaramurthy et al. | Service provider architecture for dynamically reconfigurable virtual instruments in networked environments |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
17P | Request for examination filed |
Effective date: 20150907 |
|
AC | Divisional application: reference to earlier application |
Ref document number: 2929341 Country of ref document: EP Kind code of ref document: P |
|
AK | Designated contracting states |
Kind code of ref document: A2 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
PUAL | Search report despatched |
Free format text: ORIGINAL CODE: 0009013 |
|
AK | Designated contracting states |
Kind code of ref document: A3 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: BA ME |
|
RIC1 | Information provided on ipc code assigned before grant |
Ipc: G01N 27/74 20060101AFI20160520BHEP |
|
17Q | First examination report despatched |
Effective date: 20161129 |
|
GRAP | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOSNIGR1 |
|
INTG | Intention to grant announced |
Effective date: 20170203 |
|
GRAS | Grant fee paid |
Free format text: ORIGINAL CODE: EPIDOSNIGR3 |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
AC | Divisional application: reference to earlier application |
Ref document number: 2929341 Country of ref document: EP Kind code of ref document: P |
|
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
REG | Reference to a national code |
Ref country code: GB Ref legal event code: FG4D |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: EP |
|
REG | Reference to a national code |
Ref country code: IE Ref legal event code: FG4D |
|
REG | Reference to a national code |
Ref country code: AT Ref legal event code: REF Ref document number: 910869 Country of ref document: AT Kind code of ref document: T Effective date: 20170815 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R096 Ref document number: 602014012144 Country of ref document: DE |
|
REG | Reference to a national code |
Ref country code: NL Ref legal event code: MP Effective date: 20170719 |
|
REG | Reference to a national code |
Ref country code: LT Ref legal event code: MG4D |
|
REG | Reference to a national code |
Ref country code: AT Ref legal event code: MK05 Ref document number: 910869 Country of ref document: AT Kind code of ref document: T Effective date: 20170719 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: NL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: LT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: SE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: NO Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20171019 Ref country code: HR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: FI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: AT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20171119 Ref country code: ES Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: PL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: RS Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: BG Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20171019 Ref country code: LV Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: GR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20171020 |
|
REG | Reference to a national code |
Ref country code: DE Ref legal event code: R119 Ref document number: 602014012144 Country of ref document: DE |
|
REG | Reference to a national code |
Ref country code: CH Ref legal event code: PL |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: LI Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20170831 Ref country code: MC Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: CH Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20170831 Ref country code: CZ Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: DK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: RO Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 |
|
PLBE | No opposition filed within time limit |
Free format text: ORIGINAL CODE: 0009261 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: ST Effective date: 20180430 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT |
|
REG | Reference to a national code |
Ref country code: IE Ref legal event code: MM4A |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: EE Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: IT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: SM Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: SK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 |
|
REG | Reference to a national code |
Ref country code: BE Ref legal event code: MM Effective date: 20170831 |
|
26N | No opposition filed |
Effective date: 20180420 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: LU Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20170829 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: IE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20170829 Ref country code: DE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20180301 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: SI Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: BE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20170831 Ref country code: FR Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20170919 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: MT Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20170829 |
|
GBPC | Gb: european patent ceased through non-payment of renewal fee |
Effective date: 20180829 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: HU Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT; INVALID AB INITIO Effective date: 20140829 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: CY Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 Ref country code: GB Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20180829 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: MK Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: TR Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: PT Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: AL Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT Effective date: 20170719 |